...
首页> 外文期刊>International Journal of Innovative Computing Information and Control >APPLICATION OF Q CHARTS FOR SHORT-RUN AUTOCORRELATED DATA
【24h】

APPLICATION OF Q CHARTS FOR SHORT-RUN AUTOCORRELATED DATA

机译:Q图在短时自动相关数据中的应用

获取原文
获取原文并翻译 | 示例
           

摘要

Traditional control charts are designed for processes where data are independent and identically distributed and where large historical data sets are available before a production run for estimating process parameters and computing control limits. Many processes, particularly in semiconductor manufacturing, often involve limited data sets, which result from high-mix low-volume production; in many cases, data are correlated. Therefore, charting techniques for treating short-run autocorrelated data are required. In this paper, we assess the effectiveness of control charts based on Q statistics by applying them to actual data obtained from a horizontal low-pressure chemical vapor deposition process used in semiconductor manufacturing. Our results show that Q charts enable the plotting of different types of data on the same chart, and that Q charts can detect real anomalies in data. Furthermore, we show that both Q statistics and Q statistics applied to the residuals of a time series model are practical, useful methods for the processes employed in semiconductor manufacturing.
机译:传统的控制图设计用于以下过程:数据独立且分布均匀,并且在生产运行之前可以使用大量历史数据集以估计过程参数和计算控制极限。许多过程,特别是在半导体制造中,通常涉及有限的数据集,这是由于高混合小批量生产而产生的。在许多情况下,数据是相关的。因此,需要用于处理短期自相关数据的图表技术。在本文中,我们将基于Q统计量的控制图应用于从半导体制造中使用的水平低压化学气相沉积工艺获得的实际数据中,从而评估控制图的有效性。我们的结果表明,Q图表可以在同一图表上绘制不同类型的数据,并且Q图表可以检测数据中的实际异常。此外,我们表明,将Q统计量和Q统计量应用于时间序列模型的残差,都是用于半导体制造过程的实用,有用的方法。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号